impact of a pediatric palliative care program
TRANSCRIPT
Pediatr Blood Cancer 2010;54:279–283
Impact of a Pediatric Palliative Care Program
Johannes Wolff, MD,1,2,3* Rhonda Robert, PhD,2 Andre Sommerer, RN,1 and Marlene Volz-Fleckenstein, MD1
INTRODUCTION
The goal of our study was to improve end-of-life care by offering
established, consistent pediatric palliative care services, including a
home-based service option. Children with complex, life-limiting
illnesses are increasingly dying at home. Dying at home has
increased significantly each year from 1989 to 2003 in the United
States [1]. In Canada, 44% of children receiving hospice care died a
home during 2002–2003 [2]. Feasibility of dying at home is
expanding to new populations. Levy et al. [3] found that end-of-life
care at home was feasible while caring for terminally ill children on
a neurosurgical unit in Toronto.
Troublesome symptoms such as feeding difficulties, gastro-
intestinal symptoms, breathing difficulty, and seizures were
managed by medications administered orally, by feeding tube, or
rectally; and non-invasive procedures were carried out by a nurse
under the direction of the physician.
Many benefits of coordinated, palliative care programs have
been described. In Florida, for example, several institutions created
a multi-institutional cooperative program, Partners in Care:
Together for Kids (PIC: TFK). The mission of the program is to
enable children and their families to access a supportive system of
care that is compassionate, comprehensive, family centered, and
culturally sensitive from the point of diagnosis, with hope for a cure,
through the provision of end-of-life care if needed. Positive
outcomes from this program include parent satisfaction and
increased access to services [4]. At the Children’s Hospital of
Philadelphia, a multidisciplinary, hospital-community palliative
care service, Pediatric Advanced Care Team (PACT), has
established as a primary goal, the honoring of preference for home
versus hospital-based care. Consequently, more high-quality
palliative care choices are being offered to children at end-of-life
and their families [5].
End-of-life care for children with cancer varies by institution and
community. Golan et al.[6] described the development of a palliative
care program affiliated with a pediatric oncology department in
Israel that led to a larger percentage of patients dying in the hospital
instead of at home. This article initiated a discussion [7] and
prompted us to describe our experience with developing an end-of-
life program in Germany, in an institution, where none had existed
previously. The St Hedwig’s Hospital, is the pediatric teaching
hospital of the Medical School at the University of Regensburg, in
Bavaria, Germany. Covering a large geographic area (60,000 square
km; 23,064 square miles) with a population of 2 million, 50 newly
diagnosed pediatric oncology patients are seen annually. Before
2002, patients were commonly hospitalized if death was antici-
pated. The emotional investment of the nursing and physician staff
was high, as was the level of allocated resources. Burnout syndrome
was common among nurses, physicians, volunteers, and even
clerical staff. A training or debriefing program [8,9] was absent.
The St Hedwig’s hospital-based end-of-life care conflicted
with the best-practice standards for palliative care in other countries
[9–12], where dying at home was reported to improve end-of-life
care in most cases [12,13]. We hypothesized that a pediatric
palliative care program which offered an alternative to hospital-
ization at the end-of-life would improve the patient’s end-of-life
experience.
METHODS
Program Development
This project followed the National Health Information Protec-
tion Regulations and fulfilled the hospital’s obligation for quality
control [14] as defined by the German law ‘‘Sozialgesetzbuch V.’’
Furthermore, the project was conducted in accordance with the
Catholic hospital’s vision and was approved by the hospital
Institutional Review Board.
Background. The question of where a child should die at home orin the hospital has been a subject of recent debate. We instituted apalliative care program with advanced end-of-life planning andhypothesized that a significant number of families would prefer thattheir child be at home rather than at a hospital at the end-of-life andthat the overall quality of care would thereby improve. Procedure.Study design was single group, retrospective parent report ofsatisfaction with a new palliative care program. Participants werethe caregivers, primarily parents, involved in the palliative careprogram. They completed a satisfaction survey, and their choice ofenvironment at the end-of-life was noted. Results. Fifty-one familiesexperienced the death of a child between 2001 and 2003. Most of
the patients had had cancer (n¼29, 57%), and the most frequentnon-oncologic patient had had cystic fibrosis. The parents (caregiver)participation rate was 69%. After a palliative care program wasinstituted, 69% of families preferred their child to be at home at theend-of-life, compared with 18% before the program was instituted(P¼0.0049 w2-test). The satisfaction with the medical serviceswas high (very good, 26; good, 4; satisfying, 2; acceptable, 3;deficient, 0; unacceptable, 0), independent of locale at time of death.Conclusion. When palliative care was offered either at hospital orhome, palliative care occurred more frequently at home. PediatrBlood Cancer 2010;54:279–283. � 2009 Wiley-Liss, Inc.
Key words: hematology/oncology; palliative care; quality of life
� 2009 Wiley-Liss, Inc.DOI 10.1002/pbc.22272Published online 14 October 2009 in Wiley InterScience(www.interscience.wiley.com)
——————Additional supporting information may be found in the online version
of this article.
1Krankenhaus Barmherzige Bruder, University of Regensburg,
Germany; 2Department of Pediatrics, The University of Texas M.D.
Anderson Cancer Center, Houston, Texas; 3Department of
Biostatistics, The University of Texas M.D. Anderson Cancer Center,
Houston, Texas
The authors do not have any conflicts of interest.
Grant sponsor: VKKK Ostbayern.
*Correspondence to: Johannes Wolff, Division of Pediatrics, Section of
Pediatric Neuro-Oncology, Unit 87, The University of Texas M.D.
Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX 77030.
E-mail: [email protected]
Received 19 March 2009; Accepted 5 August 2009
The pediatric palliative care program was designed to treat
children at end-of-life. The overall survival rate of patients treated in
Regensburg for pediatric oncology diagnoses before 2002 was 80%.
Based on the survival rate, we estimated 10 patients would die
per year.
The first step toward instituting the program involved a weekly
inpatient palliative care round. Hospitalized patients who were
believed to be at the end-of-life were discussed. A physician directed
the meeting, and a homecare nurse, child life specialist, chaplain,
and volunteers from the community contributed. Another step in
developing the program involved initiation of home-based medical
services. The needs for developing these services, which were
consistent with suggested standards of palliative care, were
identified by parents, patients, and staff. Home-based services were
provided primarily by the physician and a nurse. Another part of the
program development was to design a prepackaged medical travel
kit that met commercial airline standards for carry-on luggage. In its
final version, the travel kit contained (a) diagnostic equipment
including a stethoscope, tongue depressor, reflex hammer, otoscope,
and ophthalmoscope; (b) pen, paper, and forms for taking clinical
notes; prescription pad; and death certificates; (c) intravenous and
oral medications for pain [15–17] and constipation; (d) 50-ml
syringes for infusion pumps as well as tubing and standard infusion
solutions [18], and (e) wound-care materials, including sterile
gloves, swabs, tissue, bandages, and non-adherent wound cover-
ings.
The home-based visit consisted of the same steps as those of a
regular medical encounter, including taking the interim history and
recent social history, performing a physical examination, discussing
currently used medications, writing prescriptions, and arranging for
the follow-up visit.
Financial expenses and billing for home-based services were
addressed only late in the process. Philanthropic donations were
garnered for start-up and non-reimbursable expenses. Most of those
donations were made in response to a local press release about the
program and its financial need. Some medical services that were
provided were billed to insurance companies, but only a small
fraction of the expenses were covered. Liability insurance for the
health care professional who conducted the home visit was not
initially available. During the course of the study, the hospital
administration assumed provision of liability insurance.
Program Evaluation
For evaluating the program, input was also sought from parents
(caregivers) whose children had been treated at our institution and
had died before 2001. A mailed self-report questionnaire was
developed for this purpose. The first version contained 52 questions
and was tested in three face-to-face interviews between the principle
investigator of the study (physician) and parents whose children had
died. On the basis of the results, the final version was reduced to 10
items (Supplemental Appendix), followed by an invitation for
comments. A cover letter and questionnaire were mailed to families
whose children had died between 2001 and 2003.
RESULTS
We initially describe the patients served, followed with a
description of the home visits and their frequency. Thereafter, the
parents’ assessment of the program is discussed.
Patients
The accrual rate of patients who chose the palliative program
exceeded the previous resource calculation of 10 deaths per year.
After 2.5 years, between 2001 and 2003, the accrual had reached
20 patients per year. This was primarily because of an unexpected
need for palliative treatment for non-oncological diseases. Twenty-
nine of the 51 patients (57%) treated in the palliative care program
had had cancer: The list of primary diagnoses had included brain
tumors (31%; n¼ 16) [19], cystic fibrosis (18%; n¼ 9), other solid
tumors (14%; n¼ 7), leukemia (12%; n¼ 6), neurologic or
metabolic diseases (10%; n¼ 5), gastrointestinal or surgical
diseases (6%; n¼ 3), immunologic and/or hematologic diseases
(4%; n¼ 2), endocrine diseases (4%; n¼ 2), and cardiac diseases
(2%; n¼ 1).
Home Visits
The number of home visits increased from an average of 14
per month in 2001, to 24 per month in 2002, and to 39 per month in
2003 (Fig. 1). The number of visits per patient varied from once only
to once daily. A physician accompanied a nurse on 20% of the home
visits according to families’ wishes and nurses’ requests. The home-
based visit was more personal in style, often starting with an
invitation for coffee in the home. The home visits most commonly
consisted of educating parents and other caregivers about the
physiological process of dying, teaching them how to talk with their
child about dying, and providing them with emotional support. The
end-of-life issues, which were discussed either with or without the
patient present, depending on the patient’s age and parents’ wishes,
addressed the following questions: When will the child die? What
should be expected between now and the child’s death? What should
be done as the child is dying? How should the body be cared for after
the child dies? Typical medical procedures included open-wound
care, draining pre-placed chest tubes, placing or removing
nasogastric tubes, administering oral and/or intravenous pain
medication, and administering total parenteral nutrition.
The most common medical interventions included wound care,
pain medication, and infusion therapy. The vast majority of time was
used in addressing parents’ emotional and educational needs rather
than the children’s medical needs.
Home-based end-of-life care necessitated intermittent hospital-
ization for palliative procedures, e.g., gastric tube or central venous
Pediatr Blood Cancer DOI 10.1002/pbc
Fig. 1. Number of home visits per month. The palliative care program
grew constantly but unsteadily. A single patient whose home was far
from the hospital could result in significant peaks in the graph of the
overall workload.
280 Wolff et al.
line placement, operation to decrease intracranial pressure or tumor
compression on airway or bowel, major wound care. Others needed
24-hr nursing care, and medical facilities were located close to
patients’ homes for this purpose.
Parents’ Assessment
We mailed 78 questionnaires to the parents (legal guardians) of
the 51 children who had died (Fig. 1). Thirteen questionnaires were
returned to sender, suggesting a change in address without a
forwarding address. Of the remaining 65 questionnaires, 35 were
completed and returned (a 54% response rate). Separate question-
naires had been sent to both parents, but only one questionnaire was
obtained per child often with two different handwritings on the same
form. Among the 35 children for whom follow-up information was
available, 24 (69%) had died at home and 11 (31%) had died in the
hospital. The children had died 2 months to 5.5 years prior to the
parent(s) answering the questionnaire. Prior to the implementation
of this pediatric palliative care program, 18% of the children died at
home.
The parents’ overall satisfaction with the information received
about their child’s disease (item 1), with the medical care (item 2),
and with the palliative care was very good (item 3 of the
questionnaire, Fig. 1). Satisfaction was rated on the Likert scale,
with 1 being very good and 6 being unsatisfactory; the mean rating
was 1.6.
After evaluating end-of-life care preferences, our hypothesis was
confirmed: When given a choice between their child being at home
or at the hospital at the end-of-life, most families preferred their
child to be at home, as shown by the reduced frequency of children
dying in the hospital (Fig. 2, P¼ 0.049, w2-test). The satisfaction
with the overall quality of palliative care was excellent (Fig. 3). The
satisfaction with the medical services was high (very good, 26;
good, 4; satisfying, 2; acceptable, 3; deficient, 0; unacceptable, 0),
independent of locale at time of death. The satisfaction with the
medical services was not statistically different between parents
whose child died at home versus the hospital, although parents
whose child died at home tended to express greater satisfaction than
did those whose child died in the hospital (Fig. 4).
Parents mostly returned to work after the death of their child:
77% had a job before their child’s diagnosis of cancer compared
with 83% after their child’s death. Some parents separated from their
spouse after their child’s death: 91% of the parents lived together
before the illness of their child compared with 82.9% after their
child had died.
Item 9 (‘‘How long before your child’s death did you and the
physician talk about your child’s dying?’’) was primarily answered
with narratives. Parents stated, perhaps more frequently than the
hospital staff had anticipated, that they felt well informed about their
child’s death. One parent recalled having been told 2 years before
the child’s death that no curative option was left. The parent
disclosed having pretended to not understand, hoping to spurn the
medical team to search for more options.
Pediatr Blood Cancer DOI 10.1002/pbc
Fig. 2. The relative frequency of death in the hospital decreased
significantly with the development of the palliative care program
(P¼ 0.049, w2-test). This confirmed our hypothesis that creating a
palliative care program would increase the relative frequency of patients
dying at home.
Fig. 3. The overall evaluation of the palliative care provided by the
pediatric oncology department was excellent. Parents of deceased
children judged the care with use of the school marks system, which is
highly standardized and well understood in their culture. The most
frequently selected mark of ‘‘1’’ is the best possible and is rarely given
within the school system.
Fig. 4. Subgroup analysis of the satisfaction data: Parents whose child
died at home tended to express greater satisfaction than did those whose
child died in the hospital; this finding, however, was not statistically
significant. A similar tendency toward better marks was observed
among parents whose child had died after the palliative program had
started, but this finding was also not statistically significant. The mark
‘‘5’’ was not given.
Palliative Care 281
In response to the request for comments on the questionnaire,
parents voiced appreciation for having a choice between home- and
hospital-based end-of-life care. Most parents noted their appreci-
ation for the medical team members. Most parents of the children
with cancer expressed the desire for concurrent antiangiogenic and
palliative treatment. No differences were detected between the
parents whose child died at home and the parents whose child died in
the hospital.
The effects of the program on the staff and cost of the program
and services were not formally evaluated. Staff work satisfaction,
not specific to the palliative care program, was evaluated during the
same time. Overall work satisfaction ranged from moderately to
very satisfied, but this variance was likely linked more to other issues
than to the development of the palliative care program. For example,
numerous changes were occurring simultaneously in the depart-
ment, such as increased numbers of patients moving to a new
location, the start of academic components such as teaching, and the
opening of the clinical trials office [19] and research laboratory. In
the narrative comments of the staff work questionnaires, the
palliative care program was consistently praised for reducing
emotional burden and freeing up work capacity. Within the
complaints section, comments tended to change from emotional
overload to work overload.
DISCUSSION
By instituting a palliative care program with advanced end-of-
life planning and the providing the option for home-based palliative
care, many families elected to have their child at home rather than at
a hospital at the end-of-life. As hypothesized, when given a choice
of end-of-life environment, many chose to have their child die at
home with the assistance of a medical team that provided home-
based therapy. Liben and Goldman also reported an the increase in
frequency of end-of-life at home care when given the option [20].
Participants in the St. Hedwig home-based palliative care program,
reported equal or better quality of care as those receiving inpatient
palliative care. The group difference was not significant. This does
not mean that the home is always the best place for end-of-life care
or that patients receive better care at home than they would have in a
hospital. For some children, the nursing requirements for home care
exceed available resources [21,22].
The extent of benefit from home-based palliative care, when
feasible and preferred by the family, was not fully explored in this
study. Location of death has important implications. Parents
surveyed by Duffy et al.[23] described improved satisfaction while
caring for their child at home. Miano et al. [24], of the Children’s
Hospital, Genoa, Italy, reported feasible, cost effective medical
assistance, and reduced physical and psychological discomfort
consequent to implementation of a home-based palliative care
program for pediatric cancer patients at end-of-life. Johnston et al.
[25] found the benefit to home-based end-of-life care is as much a
family-based service as it is a service to the primary patient.
The parent survey has been modified for future application. Item
2 will be omitted: ‘‘How satisfied were you with your child’s
medical treatment prior to end-of-life care?’’ The medical treatment
failure was the reason for the child’s death, which parents
highlighted in their response. Thus, the question did not evoke
information that spawned ideas for improved end-of-life care.
The responses to item 3, ‘‘How satisfied were you with your
child’s end-of-life care, including pain management and emotional
support?’’, highlighted the need for the question to be separated into
several items, each of which will ask about satisfaction with one aspect
of end-of-life care: pain management, emotional support for the
patient, emotional support for family and caregivers, spiritual care,
education about the dying process, and the decision-making processes.
Items 5–8, regarding employment and marital status, also need
to be revised. Our method for asking about employment showed that
people returned to work after the death of a child, but that
information did not reflect well-being in the workplace. A more
helpful item about employment might read, ‘‘How does your current
career status compare with your career status before your child’s
illness?’’ The items about living with a spouse reflected an 8%
change. However, the effect of the child’s illness and death on the
marriage would need to be described in order to interpret the nature
of the change. Thus, the items about marriage might be combined:
‘‘How did your child’s illness and death affect your marriage?’’ By
revising questions, we expect to better utilize parent feedback.
Several administrative lessons were learned during the course of
this project. We initially targeted pediatric oncology patients for
palliative care services. However, the popularity of the home-based
treatment alternative resulted in our opening the resource to all
children at the end-of-life. The increase in participation made the
program more viable.
As for the program budget, most end-of-life care is not covered
by insurance in the German health care system. Our program is
dependent on philanthropic support. If fundraising is the manner in
which society wishes for these services to be compensated, then
legal and administrative structures that facilitate fundraising,
donation, and public recognition should be fostered. End-of-life
care at home appeared to be cost effective; however, quantitative
data was not collected. The cost of home care may impact perception
of value for prospective supporters.
The pediatric palliative care program at St Hedwig Hospital has
facilitated home-based palliative care services, while maintaining
equal or greater satisfaction with medical care. Obtaining parents
input after the patient dies was feasible and informative. Parent input
changed our ideas about end-of-life care and has improved our care
for dying children.
End-of-life preferences in this agrarian, predominantly Catholic
religion of Bavaria [26] had not been explored before the initiation
of this project. The project was a catalyst for change in other
hospitals with a similar need to improve their palliative care [27].
ACKNOWLEDGMENT
This project was described in much detail in German, and as such
was accepted by the University of Regensburg as a thesis for a
medical doctor degree; the thesis was printed by KeKiBi (in
German) for distribution. We thank the parents for answering our
questionnaire and for their moral support, the VKKK Ostbayern for
its financial support, Lydia Schmidt for data management, Brigitte
Wrede and Hugo Segerer for their helpful scientific comments,
Christoph Schutz and Konrad Baumgartner for their moral support
in sometimes difficult situations, and Robert Dechnert and Erika
Blumenstock for their editorial support for the German version.
REFERENCES
1. Feudtner C, Feinstein JA, Satchell M, et al. Shifting place of death
among children with complex chronic conditions in the United
States, 1989–2003. JAMA 2007;297:2725–2732.
Pediatr Blood Cancer DOI 10.1002/pbc
282 Wolff et al.
2. Widger K, Davies D, Drouin DJ, et al. Pediatric patients receiving
palliative care in Canada: results of a multicenter review. Arch
Pediatr Adolesc Med 2007;161:597–602.
3. Levy M, Duffy CM, Pollock P, et al. Home-based palliative care for
children—Part 1: The institution of a program. J Palliat Care
1990;6:11–15.
4. Knapp CA, Madden VL, Curtis CM, et al. Partners in care: together
for kids: Florida’s model of pediatric palliative care. J Palliat Med
2008;11:1212–1220.
5. Carroll JM, Santucci G, Kang TI, et al. Partners in pediatric
palliative care: a program to enhance collaboration between
hospital and community palliative care services. Am J Hosp Palliat
care 2007;24:191–195.
6. Golan H, Bielorai B, Grebler D, et al. Integration of a palliative and
terminal care center into a comprehensive pediatric oncology
department. Pediatr Blood Cancer 2008;50:949–955.
7. Wolff J, Wolff B. Quality of death in children: where is it higher—at
home or in the hospital? Pediatr Blood Cancer 2008;50:945–946.
8. Zinner SE. The Use of Pediatric Advance Directives: A Tool for
Palliative Care Physicians. Am J Hosp Palliat Care 2008;25:427–
430.
9. Sheetz MJ, Bowman MA. Pediatric palliative care: an assessment
of physicians’ confidence in skills, desire for training, and
willingness to refer for end-of-life care. Am J Hosp Palliat Care
2008;25:100–105.
10. Dangel T. The status of pediatric palliative care in Europe. Journal
Pain Symptom Manage 2002;24:160–165.
11. Dangel T. Poland: the status of pediatric palliative care. J Pain
Symptom Manage 2002;24:222–224.
12. Himelstein BP, Hilden JM, Boldt AM, et al. Pediatric palliative
care. N Eng J Med 2004;350:1752–1762.
13. Janssen G, Friedland C, Richter U, et al. Out-patient palliative care
of children with cancer and their families. Klinische Padiatrie
2004;216:183–188.
14. Zernikow B. Standards in pediatric palliative care in Europe—
recommendations of the IMPaCCT group. Kinderkrankensch-
wester 2008;27:228–232.
15. Collins JJ, Grier HE, Kinney HC, et al. Control of severe pain in
children with terminal malignancy. J Pediatr 1995;126:653–657.
16. Bruera E, MacEachern T, Ripamonti C, et al. Subcutaneous
morphine for dyspnea in cancer patients. Ann Intern Med
1993;119:906–907.
17. Allard P, Lamontagne C, Bernard P, et al. How effective are
supplementary doses of opioids for dyspnea in terminally ill cancer
patients? A randomized continuous sequential clinical trial. J Pain
Symptom Manage 1999;17:256–265.
18. Bruera E, Belzile M, Watanabe S, et al. Volume of hydration in
terminal cancer patients. Support Care Cancer 1996;4:147–150.
19. Wolff JE, Classen CF, Wagner S, et al. Subpopulations of malignant
gliomas in pediatric patients: analysis of the HIT-GBM database. J
Neurooncol 2008;87:155–164.
20. Liben S, Goldman A. Home care for children with life-threatening
illness. Journal of palliative care 1998;14:33–38.
21. Long T, Hale C, Sanderson L, et al. Evaluation of educational
preparation for cancer and palliative care nursing for children and
adolescents in England. Eur J Oncol Nurs 2008;12:65–74.
22. Collins JJ, Grier HE, Sethna NF, et al. Regional anesthesia for pain
associated with terminal pediatric malignancy. Pain 1996;65:
63–69.
23. Duffy CM, Pollock P, Levy M, et al. Home-based palliative care for
children—Part 2: The benefits of an established program. J Palliat
Care 1990;6:8–14.
24. Miano M, Manfredini L, Garaventa A, et al. Feasibility of a home
care program in a pediatric hematology and oncology department.
Results of the first year of activity at a single Institution.
Haematologica 2002;87:637–642.
25. Johnston DL, Nagel K, Friedman DL, et al. Availability and use of
palliative care and end-of-life services for pediatric oncology
patients. J Clin Oncol 2008;26:4646–4650.
26. Fuhrer M, Duroux A, Borasio GD. Pediatric palliative medicine.
MMW Fortschritte der Medizin 2008;150:30–32.
27. Fuhrer M. Munich coordination center of pediatric palliative
medicine–a model for the future?. Kinderkrankenschwester
2008;27:99–102.
Pediatr Blood Cancer DOI 10.1002/pbc
Palliative Care 283